Privacy Policy
Learn how OpenPond handles and protects your data.
Last updated: November 6, 2026
1. Introduction
This Privacy Policy explains how Lost Labs LLC, doing business as OpenPond AI ("OpenPond," "we," "us," or "our"), collects, uses, stores, and discloses information across our website, applications, APIs, SDK-connected services, hosted infrastructure, and related services. This includes Ponder Training, datasets, graders, evaluations, continual learning, model artifacts, inference, chat, and agent workflows.
We do not use your Customer Content to train OpenPond-owned, shared, or other customers' models. Training and evaluation you request for your own models are separate, as explained below. Where we process personal information on behalf of an organization, its instructions and applicable agreement govern that processing. Contact that organization about information it controls; contact us about your OpenPond account.
2. Information We Collect
We collect information you provide, information generated when you use the service, and technical information associated with your account and devices. Connected services and agent imports provide information according to the sources and permissions you configure. SDK task ingestion and optional telemetry integrations transmit the information your integration sends; installing an SDK does not itself authorize collection of all production conversations or logs. Avoid submitting secrets or personal information that your task does not require.
- Account, profile, contact, team, and workspace information
- Billing information, transaction records, and resource usage
- Customer Content: prompts, conversations, code, files, tasks, datasets, reference answers, graders, corrections, tool inputs and outputs, uploaded logs or traces, model outputs, configurations, and model weights or adapters you submit or generate through the service
- Run and deployment records, evaluation scores, training metrics, artifact metadata, and records of your instructions and approvals
- Usage statistics, performance metrics, and error reports
- Technical information such as IP address, browser type, device identifiers, network information, and approximate location inferred from those signals
- Information you send when requesting support or providing feedback
3. How We Use Information
We use information to provide, host, operate, schedule, secure, and support the service; process the imports, evaluations, training, continual-learning policies, inference requests, deployments, and agent actions you authorize; manage accounts, permissions, metering, billing, and payments; investigate issues and improve reliability and usability; detect abuse, fraud, and unauthorized access; communicate with you; and comply with law.
Where applicable law requires a legal basis, we process personal information to perform our agreement with you, comply with legal obligations, pursue legitimate interests such as security and service operation, or with your consent where required. Operational analytics do not give us permission to use Customer Content for model training.
4. Your Content and Model Training
We do not use Customer Content, including your datasets, task feedback, prompts, outputs, or customer-specific model artifacts, to train or fine-tune OpenPond-owned, shared, or other customers' models. Using chat, coding, evaluations, inference, imports, or support does not grant us that permission. We do not sell or license Customer Content for someone else to train models on it, or bypass this restriction by aggregating or de-identifying the content.
If you request training or enable a continual-learning policy, we process the selected data and feedback to train and evaluate models for you under those instructions. That is customer-directed training, not training OpenPond's own models. Adding a task, connecting an ingestion source, or saving a configuration does not by itself start training, deploy a model, or publish a dataset. An enabled recurring policy may run according to the settings you authorize.
We may use voluntary, nonconfidential product suggestions to improve the service. Task ratings, corrected answers, examples, uploaded support material, and other Customer Content are not treated as freely reusable product suggestions.
5. Service Providers, Integrations, and Sharing
We process information on our systems and use service providers for functions such as hosting, storage, model execution, payments, analytics, communications, and security. We disclose information as needed to deliver and support those functions. This does not grant OpenPond a right to train its own models on Customer Content.
When you select an external model provider, bring your own credentials, or connect another service, your requests may be sent to that service. Its terms, retention settings, and privacy practices also apply to its processing. Review those settings for sensitive workloads; this policy does not promise zero retention by every provider or integration.
Teammates and administrators may access information according to the workspace permissions you use. If you choose to publish or share a dataset, model artifact, deployment, or other content, the audience you select may access it. Publicly shared material may be copied by others; removing the original cannot recall copies they already obtained.
We may also disclose information when necessary to comply with law, protect rights or safety, investigate security or abuse incidents, or in connection with a business transfer subject to applicable privacy obligations. A transfer does not itself expand the training permission described in this policy.
6. Storage and Retention
Some information may remain on your device; hosted features store information on our systems or with service providers. Saved datasets, grader definitions, conversations, run history, and model artifacts require storage to remain available to you. OpenPond is not a blanket zero-data-retention service.
We retain information for as long as reasonably needed to provide the features you use, follow your instructions, maintain necessary security and billing records, resolve disputes, and comply with law. Retention depends on the record, feature, and applicable agreement. Deletion may not immediately remove isolated backups or records we must retain for those purposes. Retained information remains subject to the training restriction above.
Use available deletion and export controls or contact privacy@openpond.ai to request assistance. Deleting source data does not automatically remove information already incorporated into a model you trained or copies you exported or shared. Address those artifacts separately. Ending an ingestion connection prevents future authorized imports; it does not itself delete records already imported.
7. Workload Access and Security
Our systems process workloads and operational metadata to provide the service. OpenPond personnel do not intentionally access private workload contents or credentials unless you request or authorize access, or access is necessary for legal compliance, security incident response, or platform integrity. Support access is limited to what is reasonably needed to address the issue.
We use reasonable administrative, technical, and physical safeguards. No method of transmission or storage is completely secure. You are responsible for the permissions, credentials, and information you share with your team, agents, and connected services.
8. Cookies and Product Analytics
We use cookies, local storage, and similar technologies for functions such as authentication, preferences, security, and usage measurement. Product analytics help us understand navigation and service performance. You can manage cookies through your browser; disabling necessary storage may prevent sign-in or other features from working. Contact privacy@openpond.ai about privacy choices applicable to your use of the service.
9. Your Rights and Choices
Depending on your location and applicable law, you may request access, correction, deletion, or a portable copy of personal information, or object to or restrict certain processing. You may withdraw consent where processing relies on it, without affecting prior lawful processing, and may have the right to complain to your local data protection authority or appeal a request decision. Contact privacy@openpond.ai; we may need to verify your identity and authority before fulfilling a request. We will respond as required by law.
You can manage authorized integrations and workspace permissions and use available export or deletion controls. Use the unsubscribe link in marketing emails to stop those messages; necessary account, security, and billing communications may continue.
10. International Processing
Information may be processed in the United States and other countries where we or our service providers operate. Their laws may differ from those in your location. Where applicable law requires transfer safeguards or an additional processing agreement, those requirements apply. A specific processing region or enterprise arrangement must be agreed separately; it is not implied by a model or dataset selection.
11. Changes to This Policy
We may update this policy and its Last updated date. We will provide notice of material changes through the service, email, or another appropriate method, and obtain consent where required by law. A policy update alone does not authorize us to use previously submitted Customer Content to train our own or other customers' models.
12. Contact Us
Contact Lost Labs LLC, doing business as OpenPond AI, at privacy@openpond.ai for privacy inquiries or support@openpond.ai for general support. Expressly accepted enterprise agreements and data processing addenda govern the processing they cover; this policy does not replace those agreements.